US8014263B2  Crosstalk cancellation in cooperative wireless relay networks  Google Patents
Crosstalk cancellation in cooperative wireless relay networks Download PDFInfo
 Publication number
 US8014263B2 US8014263B2 US12543671 US54367109A US8014263B2 US 8014263 B2 US8014263 B2 US 8014263B2 US 12543671 US12543671 US 12543671 US 54367109 A US54367109 A US 54367109A US 8014263 B2 US8014263 B2 US 8014263B2
 Authority
 US
 Grant status
 Grant
 Patent type
 Prior art keywords
 cross
 rs
 signal
 method
 coupling channel
 Prior art date
 Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
 Expired  Fee Related, expires
Links
Images
Classifications

 H—ELECTRICITY
 H04—ELECTRIC COMMUNICATION TECHNIQUE
 H04B—TRANSMISSION
 H04B7/00—Radio transmission systems, i.e. using radiation field
 H04B7/14—Relay systems
 H04B7/15—Active relay systems
 H04B7/155—Groundbased stations
 H04B7/15592—Adapting at the relay station communication parameters for supporting cooperative relaying, i.e. transmission of the same data via direct  and relayed path

 H—ELECTRICITY
 H04—ELECTRIC COMMUNICATION TECHNIQUE
 H04B—TRANSMISSION
 H04B7/00—Radio transmission systems, i.e. using radiation field
 H04B7/14—Relay systems
 H04B7/15—Active relay systems
 H04B7/155—Groundbased stations
 H04B7/15564—Relay station antennae loop interference reduction

 H—ELECTRICITY
 H04—ELECTRIC COMMUNICATION TECHNIQUE
 H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
 H04L27/00—Modulatedcarrier systems
 H04L27/26—Systems using multifrequency codes
 H04L27/2601—Multicarrier modulation systems
Abstract
Description
This invention relates generally to wireless relay networks, and more particularly to canceling crosstalk in cooperative wireless relay stations.
In wireless networks, a base station (BS) is usually located near the center of a cell. As a result, the quality of signals received at a mobile station (MS) near an edge of the cell is reduced. This problem can be resolved by decreasing the size of the cell. However, this increases costs because the number of BSs needs to be increased to provide the same service for the same coverage area. In addition, this also may cause higher level of interference to MSs residing at the edge of neighboring cells. An alternative solution uses a less complex relay station (RS), which assists in communications between the BS and the MS. The RS can effectively increase coverage and reliability with decreased transmit power and cost.
The RS can be deployed anywhere in the cell wherein direct communication between the BS and the MS is impaired. The RS can also be deployed temporarily in areas where the number of MSs is expected to increase dramatically for a short time period, e.g., largescale public events.
The RS can use decodeandforward (DF) or amplifyandforward (AF) modes. The DF mode detects and demodulates received signals before retransmitting. The AF mode only amplifies the received signal before retransmitting.
Conventionally, the RS can use different frequencies or times to reduce interference. In a frequencyreuserelaystation (FRRS), the frequency bands for transmitting and receiving signals are identical. In a frequencyshiftedrelaystation (FSRS), the frequency bands are different. While the FRRS increases spectral efficiency, crosstalk interference becomes an issue because the transmit signal power is always greater than the receive signal power. In the FSRS, outofband leakage can cause crosstalk interference.
Crosstalk occurs when a transmitted signal interferes with a currently received signal. Crosstalk can be caused by undesired capacitive, inductive, or conductive coupling between the transmit antenna and the colocated receive antenna, or lines and circuits to which the antennas are connected. Crosstalk is often denoted as cochannel interference, and is related to adjacentchannel interference, see Nasr et al., “Performance of an echo canceller and channel estimator for onchannel repeaters in DVBT/H networks,” IEEE Trans. Broadcasting, vol. 53, no. 3, pp. 609618, September 2007, and Mazzotti et al., “Performance of an echo canceller based on pseudonoise training sequences,” Proc. 58th Annual IEEE Broadcast Symposium, October 2008.
Prior art techniques require the RS to transmit dedicated pilot signals, such as pseudonoise sequences, for estimating the coupling channel between the colocated transmit and receive antennas. The pilot signals change the existing signal structure of the physical layer, leading to incompatibility with legacy standards, and also result in interference at the receiver.
It is desired to perform crosstalk cancellation at the RS that is transparent to current wireless standard, i.e., the structure of the signals at physical layer remains the same at the BS and the MS.
The embodiments of the invention object provide a method for canceling crosstalk interference in a wireless frequencyreuserelaystation (FRRS) or a frequencyshiftedstation (FSRS) using signals that are designed according to current wireless standards.
The RS estimates a coupling channel between a transmit antenna and a colocated receive antenna from a previous transmitted signal and a currently received signal. Based on the coupling channel, the crosstalk interference at the RS can be reconstructed, and crosstalk interference can be reduced.
Relay Network
The RS operates as a frequencyreuserelaystation (FRRS), or as a frequencyshiftedrelaystation (FSRS). The RS can cooperate by using the identical frequency band for receiving and transmitting. Thus, depending on the specific cooperative scheme applied, there can be direct communication links 104 between the BS and the MS, as well as the indirect links 105 so as to achieve cooperation.
CrossTalk Interference at the RS
The RS can have multiple transmit antennas colocated with multiple receive antennas. A desired received symbol vector at the RS is s(n) in the n^{th }symbol period, for example, an orthogonal frequencydivision multiplexing (OFDM) symbol period. The transmitted symbol matrix or the leakage symbol matrix of the transmitted signal of the RS is {circumflex over (x)} (n−1), which is based on the transmitted symbols and the transmit filter at the RS in the last symbol period. The composite coupling channel vector from the transmit antenna to the receive antenna of the RS is h_{c}, and a white noise vector is w(n). The received symbol vector at the RS in the n^{th }symbol period is
where i(n)={circumflex over (X)}(n−1)h_{c }denotes the crosstalk interference to be cancelled at the RS. The invention recovers the desired signal s(n), i.e., s(n)=y(n)−(i(n)+w(n)). It is also noted that the RS introduces a delay of one symbol, OFDM symbol period. This enables the estimation of the coupling channel, as will be shown below, by providing the RS's receiver with a known interference.
CrossTalk Cancellation Based on Coupling Channel Estimation
We estimate 301 a coupling channel 310 between the transmit antenna 204 and the colocated receive antenna 205 at the RS 201. Based on the estimated coupling channel and a previous transmitted signal Tx 311 transmitted by the RS in a last symbol period, we determine 302 the crosstalk interference 320.
To cancel 303 the crosstalk, we subtract crosstalk interference 320 from the currently received signal Rx 330 to obtain a residual signal 340. Then, we process the residual signal to obtain the desired next transmit signal to transmit 305. This method can operate for either downlink transmissions from the BS to the MS, or uplink transmissions from the MS to BS. The next signal becomes the previous transmit signal in the next iteration.
Because the previous transmitted signal {circumflex over (x)}(n−1) 311 is exactly known by the RS, the signal can be used to estimate the coupling channel 310. In this way, our invention realizes the estimation of the coupling channel without transmitting pilot signals as in the prior art. Because the desired signal is unknown at the RS, the signal is regarded as noise when estimating the coupling channel. Because the transmit antenna and the receive antenna are colocated at the RS, the coupling channel is stationary over time. Therefore, we can use multiple previous transmitted symbol matrices to improve the accuracy of the estimation.
CrossTalk with OFDM Modulation
In one embodiment of our invention, we cancel crosstalk at the RS with a set M_{r}≧1 of receive antennas colocated with a set M_{t}≧1 transmit antennas. For Ksubcarrier orthogonal frequencydivision multiplexing (OFDM) modulation, the received symbol vector over the k^{th }subcarrier of the n^{th }OFDM symbol is
y(n,k)=s(n,k)+i(n,k)+w(n,k),
where
i(n,k)=H _{c}(k){circumflex over (x)}(n−1,k)
denotes the crosstalk interference, {circumflex over (x)}(n−1,k) denotes the transmitted symbol vector of the RS in the last OFDM symbol, and H_{c}(k) denotes the coupling channel of the RS over the k^{th }subcarrier.
A multipletap, timedomain coupling channel has a maximum delay of L OFDM sampling intervals, with the M_{r}×M_{t }channel matrix on the l^{th }tap H_{c,l}, 1≦l≦L−1. Thus, the crosstalk interference over the k^{th }subcarrier of the n^{th }OFDM symbol is
where
We define
and
h _{c}=(vec(H _{c,0})^{T}, vec(H _{c,l})^{T}, . . . , vec(_{H} _{c,L−1})^{T})^{T},
which denotes the composite coupling channel vector to estimate to obtain i(n,k)={circumflex over (X)}(n−1,k)h_{c}, where T is the transpose operator.
We determine the transmitted symbol matrix at the RS in the (n−1)^{th }OFDM symbol as
{circumflex over (X)}(n−1)=({circumflex over (X)}(n−1, 0)^{T} , {circumflex over (X)}(n−1,1)^{T} , . . . , {circumflex over (X)}(n−1, K−1)^{T})^{T}.
Then, the crosstalk interference in the n^{th }OFDM symbol is
Therefore, the original received symbol vector of the RS in the n^{th }OFDM symbol is
where
s(n)=(s(n,0)^{T} , s(n,1)^{T} , . . . ,s(n,K−1)^{T})^{T }
and
w(n)=(w(n, 0)^{T} , w(n, 1)^{T} , . . . , w(n, K− _{1})^{T})^{T}.
We denote the composite transmitted symbol matrix of the RS in the previous N OFDM symbols as
{circumflex over (X)} _{N}=({circumflex over (X)}(n−1)^{T} , {circumflex over (X)}(n−2)^{T} , . . . , {circumflex over (X)}(n−N)^{T})^{T},
and then the corresponding composite received symbol vector of the RS is
where i_{N}={circumflex over (X)}_{N}h_{c }denotes the crosstalk interference,
s _{N}=(s(n)^{T} ,s(n−1)^{T} , . . . ,s(n−(N−1))^{T})^{T},
and
w _{N}=(w(n)^{T} , w(n−1)^{T} , . . . , w(n−(N−1))^{T})^{T}.
CrossTalk Cancellation Based on LeastSquare Coupling Channel Estimation
Because the previous transmitted symbol matrix, {circumflex over (X)}_{N}, is known exactly at the RS, we can obtain a leastsquare (LS) estimate of the composite coupling channel vector as
can be determined iteratively to reduce the complexity of estimating coupling channels in multiple consecutive OFDM symbols.
Based on the estimated coupling channel ĥ_{c,LS}, the crosstalk interference at the RS is reconstructed as î(n)={circumflex over (X)}(n−1)ĥ_{c,LS}. We subtract î(n) from the currently received signal, and we obtain the estimated desired signal as
denotes the residual error vector after crosstalk cancellation.
Our crosstalk cancellation method based on the LS estimation of the coupling channel is applicable to the RS with decodeandforward (DF) or amplifyandforward (AF) modes, or other relay mechanism. Furthermore, the method is applicable to both an FRRS and an FSRS. When applied to an FRRS, {circumflex over (X)}(n−1) denotes the transmitted symbol matrix at the RS in the (n−1)^{th }OFDM symbol; when applied to an FSRS, {circumflex over (X)}(n−1) denotes the leakage symbol matrix of the transmitted signal in the (n−1)^{th }OFDM symbol, which can be obtained based on the transmitted signal and the outofband leakage of the transmit filter of the RS.
Coupling Channel Estimation and CrossTalk Cancellation at a DFBased RS
In one embodiment of our invention, the source station (SS) has M, (≧1) transmit antennas, which, depending on the direction of communication can be the BS or the MS. If we denote the M_{s}dimensional transmitted symbol vector of the SS over the k^{th }subcarrier of the n^{th }OFDM symbol as x(n,k), and the corresponding M_{r}×M_{s }channel matrix from the SS to the RS as H_{r}(n,k), then the desired signal vector at the RS is
s(n,k)=H _{r}(n,k)×(n,k).
If we denote s(n)=(s(n,0)^{T},s(n,1)^{T}, . . . ,s(n,K−1)^{T})^{T }as the composite received signal vector at the RS in the n^{th }OFDM symbol, then s(n)=H_{r}(n)×(n), where
H _{r}(n)=diag{H _{r}(n, 0), H _{r}(n, 1), . . . , H _{r}(n, K−1)}
denotes the frequencydomain block diagonal channel matrix between the SS and the RS.
In the case that the RS uses the DF mode, both the transmitted symbols of the SS and those of the RS have identical power, and the wireless channel from the SS to the RS is subject to Rayleigh fading. It can be shown that the composite signal matrix in the previous N OFDM symbols s_{N }is independent of the forwarded symbol matrix of the RS {circumflex over (X)}_{N}. In this case, the correlation matrix of the residual error vector after crosstalk cancellation based on the LS coupling channel estimation is
where R_{s}=E{s_{N}s_{N} ^{H}} and R_{2}=E{w_{N}w_{N} ^{H}} denote the correlation matrices of the composite desired signal and noise vectors, respectively, and the expectation E{·} is with respect to noise, random transmitted signal of the SS, and the random channel from the SS to the RS.
In our invention, R_{e}(n) is utilized to improve the detection of the desired signal at the RS, for example, by whitening the noise containing the residual error, or by performing the minimum meansquare error (MMSE) estimate of the transmitted symbol vector of the SS.
In the case that the transmitted signals of the SS are independent across different OFDM subcarriers and different OFDM symbols, the multitap timedomain channel between the SS and the RS has independent channel gains over different taps, and the M_{r}×M_{s }channel matrix over each tap has independently and identically distributed elements, it can be shown that R_{s}=M_{s}σ_{x} ^{2}σ_{h} ^{2}I_{NKM}, where σ_{x} ^{2 }denotes the constant power of the transmitted symbol over each subcarrier from a transmit antenna of the RS, σ_{h} ^{2 }denotes the average power gain of the channel from the SS to the RS, and I_{NKM}, denotes the NKM_{r}×NKM_{r }identity matrix. Further, we denote σ_{w} ^{2 }as the white noise power at the RS, and then R_{e,LS}(n) can be simplified as
R _{e,LS}(n)=(σ_{w} ^{2} +M _{s}σ_{x} ^{2}σ_{h} ^{2}){circumflex over (X)}(n−1)F ^{−1}(n)
When the correlation matrix of the composite coupling channel vector R_{h} _{ c }=E{h_{c}h_{c} ^{H}} is known, our invention further obtains an MMSE estimate of h_{c }based on the received signal in the previous N OFDM symbols as
ĥ _{c,MMSE} =[R _{h} _{ c } ^{−1} +{circumflex over (X)} _{N} ^{H}(R _{s} +R _{W})^{−1 } {circumflex over (X)} _{N}]^{−1 } {circumflex over (X)} _{N} ^{H}(R _{s} +R _{W})^{−1}□y_{N}.
The corresponding correlation matrix of the residual error vector after crosstalk cancellation is
which we use to improve the detection of the desired signal at the RS.
Similarly to LS coupling channel estimation, in the case that R_{s}=M_{s}σ_{x} ^{2}σ_{h} ^{2}I_{NKM}, and R_{w}=σ_{w} ^{2}I_{NKM}, ĥ_{c,MMSE }and R_{e,MMSE}(n) can be simplified as
and
R _{e,MMSE}(n)=(σ_{w} ^{2} +M _{s}σ_{x} ^{2}σ_{h} ^{2}){circumflex over (X)}(n−1)[(σ_{w} ^{2} +M _{s}σ_{x} ^{2}σ_{h} ^{2})R _{h} _{ c } ^{−1} +F(n)]^{−1 } {circumflex over (X)} ^{H}(n−1),
respectively, where F(n) and g(n) are defined above.
Based on the above simplification, the MMSE coupling channel estimation is performed in our invention as shown in
Simulation Results
We simulate a cooperative network in which the BS, the wireless FRRS, and the MS each have one transmit antenna and one receive antenna. OFDM modulation with 16 subcarriers is utilized for broadband transmission in this wireless network and QPSK modulation is applied over each OFDM subcarrier.
The maximum delay of the multipletap timedomain coupling channel from the transmit antenna to the receive antenna of the RS is two OFDM sampling intervals. The channel gains over the two taps are with independent Rayleigh fading, while the average powers decay with the delay exponentially with the exponent factor one.
The wireless channel between the source station, which may be the BS or the MS, and the RS is also with Rayleigh fading. Furthermore, the signaltonoise ratio (SNR) at the RS is set to be 40 dB. In the simulation, the signaltointerference ratio (SIR) is defined as the ratio of the average desired signal power to the average crosstalk interference power before crosstalk cancellation; the normalized meansquare error (MSE) after crosstalk cancellation is defined as the average power of the residual error normalized to the average power of the desired signal over each OFDM subcarrier.
The Figure shows that when the SIR is less than 0 dB, which is the usual case in practice, the crosstalk cancellation based on the LS and the MMSE coupling channel estimation achieve a similar improvement gain. When the SIR is greater than 0 dB, the MMSE coupling channel estimation achieves a higher improvement gain than the LS coupling channel estimation. When the SIR is higher than 15 dB, the crosstalk cancellation based on the MMSE coupling channel estimation still achieves an improvement gain, while the estimation based on the LS coupling channel cannot.
Claims (14)
Priority Applications (1)
Application Number  Priority Date  Filing Date  Title 

US12543671 US8014263B2 (en)  20090819  20090819  Crosstalk cancellation in cooperative wireless relay networks 
Applications Claiming Priority (2)
Application Number  Priority Date  Filing Date  Title 

US12543671 US8014263B2 (en)  20090819  20090819  Crosstalk cancellation in cooperative wireless relay networks 
JP2010165112A JP2011045048A (en)  20090819  20100722  Method of crosstalk cancellation in cooperative wireless relay networks and relay station 
Publications (2)
Publication Number  Publication Date 

US20110044158A1 true US20110044158A1 (en)  20110224 
US8014263B2 true US8014263B2 (en)  20110906 
Family
ID=43605307
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

US12543671 Expired  Fee Related US8014263B2 (en)  20090819  20090819  Crosstalk cancellation in cooperative wireless relay networks 
Country Status (2)
Country  Link 

US (1)  US8014263B2 (en) 
JP (1)  JP2011045048A (en) 
Families Citing this family (2)
Publication number  Priority date  Publication date  Assignee  Title 

US8619606B2 (en)  20100506  20131231  Qualcomm Incorporated  Data transmission via a relay station with ACK/NACK feedback 
JP2015207816A (en) *  20140417  20151119  富士通株式会社  Receiver, reception method, and wireless communication system 
Citations (7)
Publication number  Priority date  Publication date  Assignee  Title 

US20030086515A1 (en) *  19970731  20030508  Francois Trans  Channel adaptive equalization precoding system and method 
US20060045054A1 (en) *  20021017  20060302  Kuniaki Utsumi  Wireless communication system 
US20070038743A1 (en) *  20050517  20070215  Hellhake Paul R  System and method for communication in a wireless mobile adhoc network 
US20080165720A1 (en) *  20050217  20080710  Rong Hu  Method and Arrangement For Cooperative Relaying 
US20100008294A1 (en) *  20080711  20100114  Qualcomm Incorporated  Method and apparatus for using uplink control information for intercell decoding and interference cancellation 
US7702048B2 (en) *  20051115  20100420  Tensorcomm, Incorporated  Iterative interference cancellation using mixed feedback weights and stabilizing step sizes 
US7711075B2 (en) *  20051115  20100504  Tensorcomm Incorporated  Iterative interference cancellation using mixed feedback weights and stabilizing step sizes 
Patent Citations (7)
Publication number  Priority date  Publication date  Assignee  Title 

US20030086515A1 (en) *  19970731  20030508  Francois Trans  Channel adaptive equalization precoding system and method 
US20060045054A1 (en) *  20021017  20060302  Kuniaki Utsumi  Wireless communication system 
US20080165720A1 (en) *  20050217  20080710  Rong Hu  Method and Arrangement For Cooperative Relaying 
US20070038743A1 (en) *  20050517  20070215  Hellhake Paul R  System and method for communication in a wireless mobile adhoc network 
US7702048B2 (en) *  20051115  20100420  Tensorcomm, Incorporated  Iterative interference cancellation using mixed feedback weights and stabilizing step sizes 
US7711075B2 (en) *  20051115  20100504  Tensorcomm Incorporated  Iterative interference cancellation using mixed feedback weights and stabilizing step sizes 
US20100008294A1 (en) *  20080711  20100114  Qualcomm Incorporated  Method and apparatus for using uplink control information for intercell decoding and interference cancellation 
NonPatent Citations (2)
Title 

Mazzotti et al., "Performance of an echo canceller based on pseudonoise training sequences," Proc. 58th Annual IEEE Broadcast Symposium, Oct. 2008. 
Nasr et al., "Performance of an echo canceller and channel estimator for onchannel repeaters in DVBT/H networks," IEEE Trans. Broadcasting, vol. 53, No. 3, pp. 609618, Sep. 2007. 
Also Published As
Publication number  Publication date  Type 

US20110044158A1 (en)  20110224  application 
JP2011045048A (en)  20110303  application 
Similar Documents
Publication  Publication Date  Title 

Riihonen et al.  Optimized gain control for singlefrequency relaying with loop interference  
US20030099306A1 (en)  Method and apparatus for channel estimation using plural channels  
US20080081671A1 (en)  Mobile assisted downlink beamforming with antenna weight feedback  
US20110306291A1 (en)  Enabling downlink transparent relay in a wireless communications network  
US6442222B1 (en)  Method for error compensation in an OFDM system with diversity  
US7151755B2 (en)  Method and system for multicell interference reduction in a wireless communication system  
US7486740B2 (en)  Calibration of transmit and receive chains in a MIMO communication system  
US20120300680A1 (en)  Transmission schemes for relay  
US20100222063A1 (en)  Cellular mobile communication system, base station transmission device and mobile station reception device in cellular mobile communication system, and base station selection control method in cellular mobile communication system  
US20040146024A1 (en)  Method and system for interference reduction in a wireless communication network using a joint detector  
US8218697B2 (en)  Iterative interference cancellation for MIMOOFDM receivers  
Riihonen et al.  Spatial loop interference suppression in fullduplex MIMO relays  
US7061854B2 (en)  Efficient OFDM communications with interference immunity  
US20090219838A1 (en)  Closedloop mimo systems and methods  
US8155595B2 (en)  Method for iterative interference cancellation for cochannel multicarrier and narrowband systems  
US6711412B1 (en)  Interference mitigation in wireless communications by training of interfering signals  
US20110200126A1 (en)  Channel estimation and data detection in a wireless communication system in the presence of intercell interference  
US20100027608A1 (en)  Scrambled multicarrier transmission  
US20090061766A1 (en)  Apparatus and method for channel estimation and echo cancellation in a wireless repeater  
US7231184B2 (en)  Low overhead transmit channel estimation  
US20040192395A1 (en)  Cochannel wireless communication methods and systems using nonsymmetrical alphabets  
US20070297528A1 (en)  A High Diversity TimeSpace Coding and Decoding for MIMO Systems  
US20070201536A1 (en)  Apparatus, system and method for providing a multiple input/multiple output (MIMO) channel interface  
US20100278288A1 (en)  Channel estimation method and system for intercarrier interferencelimited wireless communication network  
US20080119190A1 (en)  Cochannel wireless communication methods and systems using relayed wireless communications 
Legal Events
Date  Code  Title  Description 

AS  Assignment 
Owner name: MITSUBISHI ELECTRIC RESEARCH LABORATORIES, INC., M Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:TAO, ZHIFENG;ORLIK, PHILIP V.;MA, JUN;AND OTHERS;SIGNINGDATES FROM 20090923 TO 20091005;REEL/FRAME:023988/0030 

REMI  Maintenance fee reminder mailed  
LAPS  Lapse for failure to pay maintenance fees  
FP  Expired due to failure to pay maintenance fee 
Effective date: 20150906 